Fast Dual Variational Inference for Non-Conjugate Latent Gaussian Models.
Mohammad Emtiyaz KhanAleksandr Y. AravkinMichael P. FriedlanderMatthias W. SeegerPublished in: ICML (3) (2013)
Keyphrases
- variational inference
- gaussian model
- mixture model
- latent variables
- factor graphs
- posterior distribution
- bayesian inference
- probabilistic model
- probabilistic graphical models
- topic models
- gaussian mixture model
- expectation maximization
- gaussian process
- gaussian distribution
- em algorithm
- variational methods
- latent dirichlet allocation
- exact inference
- generative model
- closed form
- approximate inference
- graphical models
- conditional independence
- maximum likelihood
- message passing
- belief propagation
- image intensity
- random variables
- unsupervised learning
- skin color
- model selection
- bayesian framework
- language model
- supervised learning
- multiscale